Wall wart chargers as dc to dc converters

[100volts+]

What is the best way to hook it into the main leads? I have a few 8.4 volt wall wart chargers and a few t6 cree bicycle lights. My bicycle is 30s (125.3v) hot off the charger. If I touch the prongs on the charger (8.4 v at 350ma) to the main leads the cree light works but not at full power. I read somewhere those 1600 lumen crees require at least 1 amp at 8.4 volts so I am going to order an 8.4 1 amp charger but I still am trying to figure out the best way to hook it into the main line

 

[amberwolf]

Best? Depends on what qualifications make it "Best" for you.

If you need it to be it's most efficient/etc., electrically, then you need to open it up and wire the input to a point on the electronics side of the diode(s) or bridge rectifier, rather than the AC input side. That gets you past the diode voltage drops, and the current limits of them and any heat they produce under load.

If you need it to stay sealed (assuming it started life that way), then you might try a high-wattage (80W+) soldering iron and see if you can tin the ends of the AC prongs, then solder the input wires to those. To prevent plastic melting at base of prongs, use some stats or whatever to clip some copper or aluminum strip to the prong as close to the base as you can, away from teh tip. If you can't do that, wrap a single piece of kleenex or toilet paper or cottonball around that base and add a drop or two of water, and hold it there with needlenose whileyou solder. Either way, don't keep the iron on there any longer than needed to start the solder melting and wicking onto the prongs.

If you have the space you could use an actual wall outlet in your battery box, wired to your pack's output, and just plug the wallwart into it.

 

[Barnfresh]

Please keep us posted here on what you come up with. I'm at 100.8v hot off the charger and was going to try a Sony or Dell 110v ac laptop charger.

 

[100volts+]

I looked at the laptop chargers but they put out 17 to 19 volts which is too much for that light that I use

 

[mfj197]

Unless wall power supplies have changed since I looked at them (which is possible) they will not work as DC-DC converters at all. The main voltage conversion is done by differing windings of a transformer, which will only work with AC. If you put DC on the input it will get hot but not output anything at all I'm afraid.

AC voltage conversion is easy and efficient, hence mains voltage is distributed via AC. DC voltage conversion is rather more challenging.

Michael

 

[100volts+]

as you can see in the picture above the wall wart does work because that is how the headlight is powrded in the picture I have the main leads of my bike connected to the prongs on the wall wartand and the other end is connected to my bike light which is on .

 

[cycborg]

Unless wall power supplies have changed since I looked at them (which is possible) they will not work as DC-DC converters at all. The main voltage conversion is done by differing windings of a transformer, which will only work with AC. If you put DC on the input it will get hot but not output anything at all I'm afraid.

AC voltage conversion is easy and efficient, hence mains voltage is distributed via AC. DC voltage conversion is rather more challenging.

Michael

They have in fact changed substantially since you looked at them. The old topology was:

transformer -> rectifier -> low-pass filter -> linear regulator

But for several years now the topology has been:

rectifier -> switching regulator

So the input to the regulator is now the full-amplitude rectified AC. The regulator handles this just fine despite the large amplitude and full-amplitude ripple, not to mention the fact that the amplitude can be either US-standard 120 VRMS or EU-standard 240 VRMS.

Compared to these demands, 48 or 72 or 1xx VDC input is a walk in the park! As amberwolf notes, you can even crack the box open and bypass the rectifier if you want to eke out a few more percent efficiency.

Only trick is you have to get the polarity right. When I did this, I tried both polarities - one worked and one didn't, and the latter didn't cause any damage.

Edit: check out this teardown of the iPhone charger. 100-240 VAC to 5 VDC @ 1 A, in a cube about 2 cm on a side and weighing probably a few tens of grams. Astonishing. But this is just a refinement of how things have been done for a while now.

 

[dnmun]

you don't have to guess about the polarity of the DC input when you attach to the input caps because they are labeled positive + and negative - so just attach you input power leads with the correct polarity since reversing polarity on the caps and other devices will damage them. you should not use the spark test to determine how to connect wires where there is an active device and electric circuit.

the value of bypassing the rectifier diode bridge is in the ability to carry more current to the input than can be carried through the 1/2 of the rectifier diode bridge used if you attach the DC to the normal AC input plug.

but in the case where you attach the DC to the AC input and it goes through the rectifier, then the polarity of the DC input doesn't matter. the 1/2 of the rectifier with correct polarity for that DC input will carry the power.

 

[cycborg]

where you attach the DC to the AC input and it goes through the rectifier, then the polarity of the DC input doesn't matter.

As it happens, this isn't the case with the particular adapter I'm using. Not sure why, and I'm not really motivated to open 'er up and poke around, but it is what it is.

 

[teklektik]

I looked at the laptop chargers but they put out 17 to 19 volts which is too much for that light that I use

You haven't given any details of the light you are using but at the end of the day, LEDs really want a specific current. Quality/high output LED lights frequently have internal current limiting so the current is constant with varying battery voltage. On the other hand, cheapy hand flashlights typically skip the current regulator and just match the LED characteristic more or less to the battery voltage

So, since you are talking about a 1600LM light, I'm thinking that you have a single T6 LED (roughly 2.9V-3.5V operating range) with an internal current regulator to delivers a constant 1000ma or so and that isn't too fussy about the input voltage.

If this is true, then you should be able to get by with a 9V power adapter with a 1.5A or 2A rating. A quick eBay search reveals many candidates. Here you just rely on the existing internal current regulator to work normally - it will just think the battery is HOC.

As far as the mains hookup is concerned:

• I find that it's easy to solder input leads to the prongs right up against the case (hot iron with lots of heat capacity so it goes quickly). We are only talking about 150ma or so so you can use fairly thin wire (eg 20-22ga) that will solder easily and quickly. Then lop off the protruding prongs. Run the wires flat along the adapter bottom and put a blob/bead of ShoeGoo or E6000 glue over the prong stubs (as insulation) and wires (gluing them to the adapter bottom as a strain relief). Fitting the in/out leads with connectors makes a nice little package.
  
  Bypassing the input bridge only lowers the minimum operating voltage a bit and may compromise operation as others have mentioned. These adapters will typically run with an input down to 45VDC or so where the extra 1.5V of input headroom from the bridge bypass may be useful for low voltage batteries. But you are running 125V so there is no electrical advantage to doing surgery on your unit. My 2 cents...

 

[amberwolf]

where you attach the DC to the AC input and it goes through the rectifier, then the polarity of the DC input doesn't matter.

As it happens, this isn't the case with the particular adapter I'm using. Not sure why, and I'm not really motivated to open 'er up and poke around, but it is what it is.

I have a larger power supply for (I think) an Xbox, and it too has this issue; but one step worse: It seems to need both polarities in order to to work, though I have so far not had time and equipment at the same time to diagnose it and figure out exactly why.

I've had several small wallwarts that are the same as with yours, only one polarity works if it's DC. I'm figuring it's probably only a partial rectifier and doesn't actually use the whole AC waveform. I haven't opened them up but since they also work.

 

[mfj197]

They have in fact changed substantially since you looked at them. The old topology was:

transformer -> rectifier -> low-pass filter -> linear regulator

But for several years now the topology has been:

rectifier -> switching regulator

So the input to the regulator is now the full-amplitude rectified AC. The regulator handles this just fine despite the large amplitude and full-amplitude ripple, not to mention the fact that the amplitude can be either US-standard 120 VRMS or EU-standard 240 VRMS.

Compared to these demands, 48 or 72 or 1xx VDC input is a walk in the park! As amberwolf notes, you can even crack the box open and bypass the rectifier if you want to eke out a few more percent efficiency.

Only trick is you have to get the polarity right. When I did this, I tried both polarities - one worked and one didn't, and the latter didn't cause any damage.

Edit: check out this teardown of the iPhone charger. 100-240 VAC to 5 VDC @ 1 A, in a cube about 2 cm on a side and weighing probably a few tens of grams. Astonishing. But this is just a refinement of how things have been done for a while now.

Cycborg, thank you for that useful update. Dear me - shows how long ago I was messing around with wall power supplies!

Anyway, one area I'm rather more current is on LED technology. I'm guessing the OP's light uses a Cree XM-L emitter (T6 is only the 'bin' classification but doesn't tell you which LED it is). XM-L emitters can take up to 3A, for which it needs a nominal 3.35V and therefore consuming almost exactly 10W of power. Assuming 90% efficiency (good drivers will get this) he will need 11.17W of input power, i.e. If using a 9v wall wart it must produce 1.25A on the output. A 12V one would work with 1A output.

I'm going to be using an RC battery eliminator circuit to provide the 7V required for driving my home-made twin XM-L setup, but then I'm only using a 36V drive battery so am not having to do the voltage reduction you guys are.

*Now goes off to find the nearest wall wart to dismantle... *

Michael

 

[amberwolf]

*Now goes off to find the nearest wall wart to dismantle... *

If you have them around, Goodwill and Savers often have entire wall sections full of them for a buck or less (and frequent 50% off days).

Almost any lightweight wallwart marked as "100-240VAC input" seems to work on a lot lower DC voltage than that (though not all will, and some of the older Apple powerbook adapters don't seem to work on DC input at all, or require at least 90-100VDC+ to startup (though some will run on lower voltages after they've started).

 

[Barnfresh]

I looked at the laptop chargers but they put out 17 to 19 volts which is too much for that light that I use

How about these, http://www.powerstream.com/dda.htm

Would it power something like this ok; http://www.amazon.com/1200LM-XML-T6-Headlamp-Headlight-Bicycle/dp/B00GFLQMAK/ref=pd_sim_sg_5?ie=UTF8&refRID=1MYYMEEGH9HS90P0D2W4

 

[100volts+]

Lots of option's on powerstream but a little pricey for me.

 

[100volts+]

Lots of option's on powerstream but a little pricey for me. Then again sometimes you get what you pay for.

 

[dnmun]

*Now goes off to find the nearest wall wart to dismantle... *

If you have them around, Goodwill and Savers often have entire wall sections full of them for a buck or less (and frequent 50% off days).

Almost any lightweight wallwart marked as "100-240VAC input" seems to work on a lot lower DC voltage than that (though not all will, and some of the older Apple powerbook adapters don't seem to work on DC input at all, or require at least 90-100VDC+ to startup (though some will run on lower voltages after they've started).

i just found a pile in a free box on the street. was able to put 4 in my pockets.